FI120116B - Information management techniques for metabolic related data - Google Patents

Description

Information management techniques for metabolic related data

Background of the Invention

The invention relates to information management techniques for metabolic-related data, such as data related to lipids and / or other molecular classes sharing common moieties, such as glycans.

Lipids are an important and highly diverse class of metabolic products with structural, energy storage, and communication roles. Lipid metabolism is recognized to play a key role in several diseases, such as arteriosclerosis, diabetes, and Alzheimer's, to name but a few. In spite of this importance, bioinformatics strategies that fully utilize modern analytical and informatics technologies have not yet been proposed.

Lipids are a diverse class of biological molecules that play a key role as biological cell membranes, energy stores, and messenger molecules. Lipid metabolism disorders are associated with a number of human diseases, including diabetes, Alzheimer's disease, arteriosclerosis and infectious diseases. While research on lipids and metabolites has generally been overshadowed by genomic developments in recent decades, recent renewed and rapidly growing interest in lipids, which has triggered several new lipid research projects, illustrates their critical importance in biology. Lipidomics as a field aims to characterize species of lipid molecules and their biological roles in relation to the expression and function of proteins involved in lipid metabolism, including gene regulation.

There are several public resources that represent various aspects of lipid information, such as LIPID MAPS, Lipid Bank, LIPIDAT, CyberLipids, and Lipid Base. New alliances have been formed, such as LIPID MAPS (Lipid Metabolites and Pathway Strategy), and other pioneer groups in Europe and Japan are working towards the same goals. The LIPID MAPS Consortium introduced 30 nomenclatures that allow the presentation of a lipid compound under a unique 12-digit identifier. Following the same classification and nomenclature system proposed by the LIPID MAPS Consortium, the JCBL (Japanese Conference on Biochemistry) of Lipids maintains a related database, Lipid Base, which also maintains MS / MS fragment information from individual lipid species. Recent developments in the field of analytical methods for lipid analysis, particularly liquid chromatography coupled to mass spectrometry (LC / MS), require extensive lipid libraries to enable systemic identification of lipids, findings and further research, integrated studies which combine multi-tissue lipidomics profiles with other levels of biological information, such as gene expression and proteomics, have become possible thanks to such properties.

Because of the high amount of information available due to lipid-rich data-intensive experiments, the database system must be effectively connected to an analytical platform for lipid profile data, chemo-10 and bioinformatics to identify compounds and connect information to other levels of the biological organization.

The diversity of lipids in different organisms, tissues and cell types results in the vast majority of the relevant lipids not being identified, and no single database can hardly cover all possible lipids. Thus, there is a need for a mechanism that would facilitate the discovery of new lipid species among available data in biological systems.

In addition, the currently available lipid pathway level representation in databases such as KEGG is limited to generic lipid pathway path representations, i.e., they mainly contain major group information, and override fatty acid side chain information, and as such lack the level of detail available in LC / MS. Existing databases are really very useful with regard to the automatic identification of very large peaks. Thus, there is a need for ways to identify individual molecular species. One related problem is how individual molecular species are linked to known metabolites. For example, current path chart data bases contain information only on generic lipid classes such as path charts, including phosphocholine. But there is no information about the fatty acids below. As a result, there may be hundreds of different species of phosphocholine. Another problem related to this is how to develop the diversity of lipid compounds using information technology.

BRIEF DESCRIPTION OF THE INVENTION j

It is therefore an object of the invention to provide a method, hardware and software product for alleviating at least one of the above problems. The object of the invention 35 is achieved by a method, hardware and software products which are characterized by what is stated in the appended independent claims. The dependent claims disclose specific embodiments of the invention.

One aspect of the invention is a method of processing information on compounds of molecular classes which share common moieties. Method 5 comprises the steps of: maintaining flowchart information about the compounds at the level of the individual compound and / or at the generic class level; developing a diversity of compounds based on a set of seed structures, each representing a higher than average probability of occurring in the lipid compound; using a formal description language to express seed structures; using the structural elements to generate the expected spectra for each compound using known experimental conditions for mass spectrometry; Performing one or more spectroscopy experiments to obtain compound information; and combining the resulting compound information with existing information on molecular classes.

In a representative embodiment of the invention, the compounds of the molecular classes sharing common structural moieties comprise lipids, and the invention is hereinafter described with reference to lipids.

The step of maintaining path diagram information about lipid compounds is known per se. There are several bio-informatics approaches to maintaining path diagram information, and the detailed description is omitted.

According to one embodiment of the invention, the formal markup language is SMILES, which is an acronym for Simplified Molecular Input Line Entry System. The construction of a particular lipid class can be based on a SMILES diagram for that class. First, a generic SMILES diagram is developed, for example by hand. The length of the fatty acid chain is then varied to create several or all possible compounds of this class within a particular fatty acid chain; the length of the window. For example, the PERL parser can be developed to vary the length of the fatty acid chain. Once a compound SMILES representation has been developed, the i SMILES representation can be converted to a canonical (= unique) SMILES representation.

Another interesting feature of this method relates to its ability to generate 35 systematic names algorithmically.

4

In the step of using the structural elements to generate the expected spectra for each compound, it is preferable to use commonly used experimental conditions.

Existing lipid information includes path diagram information at the individual compound level and / or at the generic class level. Alternatively, or in addition, existing lipid information may include co-regulatory information with other compounds between different biological samples.

According to one embodiment, the method further comprises linking information from the individual compound to information at other levels.

Information at other levels may include information about proteins or genes involved in the metabolism or biological variation of a single compound.

According to one embodiment, the method further comprises using information at the level of the individual compounds and their variation within a particular portion of the organism in different biological samples to find inter-compound dependencies between different parts of the organism.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described in more detail in connection with preferred embodiments, with reference to the accompanying drawings, in which: Figure 1 illustrates the principle of the invention;

Figure 2 shows a database diagram for the presentation of lipids;

Figure 3 shows a method for the systematic construction of glycerophospholipids;

Figure 4 illustrates a technique for displaying structures of lipid compounds using SMiLES;

Figure 5 is a graph of SMILES showing the core variables of fatty acids;

Figure 6 shows the structures of glycerophospholipids;

Figure 7 shows an exemplary nomenclature of glycerophospholipids | l for 30 classes; {t

Figure 8 illustrates a technique for using structured names in a linking step; !

Figures 9A and 9B, which form a logical drawing of one embodiment

are algorithmically generated SMILES for one of the precursors of the fatty acid chains

35 Character Set; I

Ϊ i f f 5

Figures 10A and 10B show the development of characteristic MS / MS spectra for individual species;

Figure 11 shows a scoring system;

Figure 12 illustrates how tissue-based study of lipid profiles at the level of the individual molecule can reveal interdependencies between biological processes in different compartments of organisms; and

Figure 13 summarizes the steps of the method of the invention.

DETAILED DESCRIPTION OF THE INVENTION

Figure 1 illustrates the working principle of the invention. Reference numeral 100 generally depicts the architecture of the data system in which the invention may be practiced. Lipid database 102 is accessible to database management block 104 and spectroscopy software block 106 via processing block 108. In this implementation, spectroscopy software block 106 supports liquid chromatography / mass spectrometry, but other techniques may be used. The primary added value of a data system 15,100 over known systems is two-fold.

Database management block 104 provides better lipid pathway reconstruction and spectroscopy software block 106 for improved interpretation of lipid profiles.

Further details of forming a suitable database management block 104 are described in the same applicant's Finnish patent application FI20055198, entitled "Visualization Technology for Biological Information", filed April 28, 1995. One aspect of the present invention disclosed by the same applicant is steps: 1) establishing a user interface and receiving the biological information query via the established user interface, 2) maintaining connections to multiple databases storing at least part of the non-overlapping biological information, 3) determining which database contains the biological information associated with the received query; ) sending a database query to the specified database 30 and receiving the result of a database query comprising biological and / or chemical and biological relationships between biological and / or chemical entities; 5) establishing a network based on the result of a database query, wherein the step of creating a network comprises mapping biological and / or chemical entities to network nodes and mapping relationships to network connections; 6) 35 defining a distance matrix for assigning a distance to a plurality of pairs of network nodes, wherein each distance is computed over a plurality of dimensions; 7) applying the dy 6 reduction function to describe the distance matrix for a smaller number of dimensions; 8) searching the neighbors of the selected network node based on the distance matrix to clarify the biological role of the selected network node; 9) adjusting the dimensionality reduction function based on one or more research contexts of biological and / or chemical information to target the study toward the relevant focal point; and 10) reconstructing and visualizing the network based on the adjusted dimension reduction function.

The method may further comprise describing each of the one or more research contexts to a network node. The imaging step may comprise viewing the process from multiple dimensions to two dimensions. The distance function may be based on network topology and / or relationships due to experimental data. Such relationships resulting from experimental data may include a measure of correlation.

Said beams for the formation of a suitable spectroscopy software block 106 are described in the same applicant's Finnish patent applications 15 FI20055252, FI20055253 and FI20055254, all entitled "Analytical Techniques for Liquid Chromatography / Mass Spectrometry", filed May 26, 2005.

One aspect of the invention disclosed in said co-applicant application F120055252 is a method for analyzing liquid chromatography / mass spectrometry data [= "LC / MS data"], comprising: 1) preparing a plurality of sample runs; 2) processing each prepared sample run on an LC / MS spectrometer to obtain a spectrum for each processed sample run; 3) internally presenting each spectrum as a mass / charge ratio plotted against the retention time; 4) performing a first peak detection to detect peaks in each spectrum; 5) generating a plurality of aligned f 25 spectra internally by aligning the detected peaks of each spectrum; and 6) j performing a second peak detection in the first peak detection for detecting undetected peaks, wherein the second peak detection comprises using knowledge of other spectra in the plurality of targeted spectra. |

One aspect of the invention disclosed by the same applicant application FI20055253 is a method for analyzing LC / MS data, comprising: 1) generating a plurality of sample runs; 2) processing each prepared sample run on an LC / MS spectrometer to obtain a spectrum for each processed sample run; 3) each spectrum is plotted as a mass / charge ratio versus retention time; 4) performing a first peak detection 35 to detect peaks in each spectrum; 5) visualizing the peaks of each spectrum, wherein the visualization step comprises 5a) depicting each peak to be visualized in a 7 coordinate system in which the first coordinate indicates the mass / charge ratio and the second coordinate indicates the retention time; and 5b) assigning a specific visual attribute to each peak to be visualized.

One aspect of the invention disclosed in the same Applicant application FI20055254 is a method for analyzing LC / MS data, comprising: 1) generating a plurality of sample runs; 2) processing each prepared sample run on an IC / MS spectrometer to obtain a spectrum for each processed sample run; 3) internally depicting each spectrum as a mass / charge relationship relative to retention time; 4) performing a first hui-10 red detection to detect peaks in each spectrum; and 5) finding the peak of the analyte peak closest to the analyte peak and normalizing the peak to be analyzed based on the distance between the peak of the analyte and said nearest standard compound peak.

LC / MS data analysis techniques can be further enhanced by additional features. For example, the second peak expression may comprise detecting local maxima and / or recursive threshold detection. The method may further comprise normalizing spectra, for example, injecting one or more standard compounds at a predetermined concentration for each sample run prior to the processing step to obtain a set of peaks of standard compounds for each standard compound injected. The method may further comprise searching for the peak of the nearest standard compound to be analyzed and normalizing the peak to be analyzed based on the distance between the peak of the analyte to be analyzed and the peak of said nearest standard compound. The alignment step may comprise the steps of: generating peaks from each spectrum, creating a peak list, and for each peak of each peak list, finding the corresponding peak of the peak list using a predetermined distance measure. The distance measure may be based on the weighted combination of \ m / zp - m / zm \\ a \ rtp -rtm I, where m / zp and rtp are the peak charge-to-charge ratio and the retention time of a single peak list, respectively, and m / z, -30 means the average mass-to-charge ratio and the retention time for all peaks from different peaks, divided into the same row of the peaks main list. The method may further comprise visualizing the peaks of each spectrum, wherein the visualization step comprises: describing each peak to be visualized in a coordinate system, wherein the first coordinate indicates the mass-charge ratio and the second coordinate indicates the retention time; and assigning to each vertex to be visualized a specific visual attribute. The visualization method may further comprise visualizing 8 peaks from the first group and the second group of samples, and the specific visual attribute is based on the ratio of average intensities between the corresponding peaks in the first group and the second group. Still further, the visualization method may comprise visualizing peaks from a group of samples, and the specific visual attribute is based on the variation of peak intensities within the group of samples.

Figure 2 is an illustrative database diagram for the presentation of lipids. In this illustrative but non-limiting embodiment, the lipid data is stored in a native XML database implemented with 10 Tamino XML Server. The entry of each compound in the database includes information about the internal identifier, scoring information, class, canonical SMILES, the molecular formula, molecular weight, and isotopic distribution. PERL scripts can be used to convert data into XML documents. The resulting XML documents are loaded into the Tamino database mass background tool. 15 XMLSPY and Tamino Schema Editor Software, respectively, can be used to construct and validate logical and physical diagrams. Tamino XML Server and Tamino Schema Editor Software are available from Software AG in Germany and XMLSPY software from Genuine lnc.

Lipids are a diverse group of molecular species, broadly defined as hydrophobic or amphipathic small molecules, which may be formed wholly or in part by carbanion-based condensation of thioesters and / or by carbonation-based condensation of isoprene units. The focus of this description of particular embodiments is primarily on the development of informatics methods for the study of glycerophospholipids, sphingolipids, glycerol lipids and sterol esters.

The major structural variant between the above classes is variation within one or more fatty acid chains forming a lipid molecule. For example, glycerophospholipids may be represented by a few major groups, such as choline or ethanolamine, while the modifications associated with the functional groups and the variety of possible fatty acid combinations are much greater.

One preferred approach to developing a diverse set of lipids, to facilitate identification in lipidomics assays, is to develop a set of "seed" fatty acids that are most likely to occur in living systems The selection of seed fatty acids disclosed herein reflects a focus on Fatty acid seeds are expressed by the SMILES (Simplified Molecular Input Line Entry System), a human-readable linear index system of atoms and bonds that is governed by precise syntax rules. While generally 5, any compound may have multiple SMILES representations, canonical versions are available that allow unique SMILES representations. The embodiments described herein can be implemented with a Daylight canonical SMILES presentation (Daylight, Chemical information system, Inc.) The SMILES representations are algorithmically constructed for all these seed-fatty acid chains and are shown in Figures 9A and 9B. The lipid database is constructed using systematic names approved by the LIPID MAPS consortium.

Each compound in the database can be assigned a score based on the natural abundance of the fatty acids from which the compound is made. Common factors to consider when scoring are the natural abundance of fatty acids and the odd or even number of carbon atoms in the fatty acid chain. In addition, different bonds of fatty acids to the major group of lipids receive different scores. The scoring system is shown in Figure 11. The total scoring is then the product of all fatty acid points.

The construction of a particular lipid class is based on the SMILES diagram for that class.

When manually generating a SMILES generic chart, PERL parsers can be developed to vary the length of the fatty acid chain to create all possible compounds of the class in a given window of the selected fatty acid chain length.

Once SMILES has been developed, SMILES can be converted into a canonical (unique) representation of SMILES. Another interesting property of this method

25 femininity is related to its ease of generating systematic names algorithmically. I

Daylight's SMILES tool kit software can be used to develop canonical j SMILES presentations. Daylight Toolkit is tailored to provide molecular weights and accurate masses of compounds. The exact masses of the elements are taken from the standard literature. One method for the systematic construction of glycerophospholipids is summarized in Figure 3.

Figure 11 shows a scoring system. Each compound in the database is assigned a score based on the natural abundance of the fatty acids from which the compound is made. Common factors to consider when scoring are the natural run-35 saus fatty acid and the odd or even number of carbon atoms in the fatty acid chain.

In addition, different bonds of fatty acids to the major group of lipids receive different scores. The total score is then the product of all fatty acid points. Random scoring S for any lipid compound having fatty acid chains with scoring variables Vj (at position Sn1), Vj (at position Sn2) and Vk (at position Sn3). For compounds having a single fatty acid chain at position Sn1 or 5 Sn2: S = V, or Vj

For compounds having two fatty acid chains at positions Sn1 and Sn2: = VjXVj

For compounds having three fatty acid chains at positions Sn1, Sn2 and Sn3: 10 = V x Vj x Vk

Figure 3 shows a method for the systematic construction of glycerophospholipids. Step 3-1 generates a generic SMILES diagram having a structure that fits the class of glycerophospholipids. The SMILES diagram showing the fatty acid seed variables for positions Sn1 and Sn2 and for the main group 15 variable (represented by X) at position Sn3 (according to SMILES syntax rules, fatty acid seed variables are written in parentheses, represented as branched strings in Jo) is shown in Figure 6.

In step 3-2, the respective systematic names are used for each of the 20 fatty acid seed SMILES to construct a generic name chart for algorithmic development of the names. An exemplary name chart for the class of glycerophospholipids is shown in Figure 7. An exemplary name table for retrieving systematic names is shown in Figures 9A and 9B.

Step 3-3 uses a PERL script that develops compounds

all possible SMILES and their systematic names. In step 3-4, the SMILES are converted to canonical SMILES (eg using daylight using the SMILES Toolkit). Step 3-5 involves obtaining a molecular formula from SMILES and calculating the molecular weight of the resulting molecular formula. In Steps 3-6, a random score is calculated for absorbing compound abundance

sex. In steps 3-7, the isotope distribution is determined by the molar I of the compound

kyylikaavasta. In steps 3-8, the isotope distribution is tailored to the resolution of the mass spectrometer used.

Spectrum spectra can be used in combination with LC / MS based seion-35 nan. To facilitate identification of lipids, mass spectrometry data

based on this, an isotope distribution can be calculated for each compound in the database. I

! l 11 This isotope distribution may be based on the observed natural abundance of each element of the chemical formula. The masses and abundances of isotopes of a particular chemical composition are predicted using appropriate software, such as the open source Isotope Pattern Calculator. This theoretically developed distribution of Theo-5 is very useful for comparing isotope diagrams of mass spectrometry data. But the distributions obtained from a mass spectrometer depend on its resolution. The PERL script can be used to convert the calculated distribution to the desired distribution according to the resolution. Distributions can be plotted.

The following description relates to the development of diversity in lipid compounds. The fact that these fatty acid chains remain part of most lipid structures makes it possible to construct lipid classes by algorithmic means. Differences in the lengths of fatty acyl / alkyl chains and degrees of unsaturation create great diversity within a certain class. The lipid database may include major classes such as fatty acyls, glycerolipids, glycerophospholipids, sphingolipids, and sterols. The category of fatty acyls includes fatty alcohols, fatty aldehydes, fatty carboxylic acids, fatty acyl CoAs / ACPs and eicosanoids. The class of glycerolipids is quite a huge class in this database and includes subclasses such as monoacyl / alkylglycerols, diacyl / alkylglycerols and triacylglycerols. 20 The number of permutations of fat / alpha chains at the three positions of glycerol, namely sn-1, sn-2 and sn-3, makes this class of combinations extremely enormous. Glycerophospholipids are another important class that includes glycerol phosphocholines, glycerophosphoethanolamines, glycerophosphoserins, glycerophosphates, glyceropyrophosphates and glycerophosphoglycerols. These compounds contain both mono- and diacyl / alkyl glycerophospholipids. Plasma genes are a specific class of phospholipids in which the fatty acid chain of glycerol contains O-alkenyl ether (-O-ChNCH-) bonds. In one embodiment, the subclass of plasmologenes is 181548. The class of sphingolipids includes sphingoid strains, various ceramides including ceramidophosphorinositols, ceramldiphosphocholines, ceramidophosphoethanolamines, N-acyl-sphingosines, N-acyl-sphingans, and the like. Sterols contain compounds of the cholesteryl ester type.

The lipid database contains essentially all possible lipids whose fatty acid lengths (or major groups if present in the class) can be algorithmically varied. One limitation of the SMILES method is its difficulty in developing SMILES structures for more complex lipids. For example, 12 complex lipids, such as glycosphingolipids, whose SMILES structures are difficult to develop algorithmically, can be constructed manually. Another limitation of this database is redundancy. Lipids having the same composition are difficult to distinguish. The redundancy problem can be partially addressed based on the scoring values, since scoring sorts the redundant lipids based on their estimated frequency in nature. More common lipids get lower points and vice versa. It is advantageous to adjust the scoring values for different organisms. Fragmentation and chromatography libraries are needed to address redundancy considerations. Molecular-10 ionic fragments corresponding to individual molecular species, preferably produced at different ionization ratios, together with an analytical method reproduced with retention time information, give a unique characteristic of a single molecular species.

Figure 4 illustrates a technique for displaying structures of lipid compounds using SMILES. Reference numeral 400 generally describes the structure of phosphocholine (PC). Phosphocholine structure 400 contains fatty acids at positions sn-1 and sn-2, glycerol backbone and choline at position sn-3. Like many lipids, phosphocholine is a class of molecules in which fatty acids at positions sn-1 and sn-2 can be varied to develop different phosphocholine compounds. Seed fatty acids are used, including common fatty acids such as palmitic or oleic 20, etc., less common such as odd-chain fatty acids, hydroxylated fatty acids, peroxides, etc.

Figure 5 shows a SMILES diagram showing the core variables of the fatty acids at positions sn-1 and sn-2 and the major group variable (represented by X) at position sn-3. According to the syntax rules in SMILES, the seed variables of fatty acids are written in parentheses, representing them as branched chains.

Figure 6 shows the structures of glycerophospholipids having major groups such as phosphocholine (PC), phosphoethanolamine (PE), phosphoserine (PS), phosphoglycerol (PG), phosphoinositol (P1), phosphate (PA) and pyrophosphate (PPA).

Figure 7 shows an exemplary nomenclature for a class of glycerophospholipids.

Figure 8 illustrates a technique for using structured names in the linking step. Generally speaking, the main functional group defines the class of lipids. Transformation between the various classes or their intermediate levels occurs at the level of the functional 35, while structural elements, such as fatty acids specific for a single lipid species within the lipid class, are retained.

13

Figures 9A and 9B, which form one logic drawing, show algorithmically generated SMILES for a set of precursors of fatty acid chains.

Figures 10A and 10B show the development of characteristic MS / MS spectra for individual species. Using the full-scan MS method followed by chromatography, the parent ion and retention time are recorded. Parent ion fragmentation, using MS / MS or similar methods, produces ion fragments which, together with MS information and retention time, help to determine a single compound.

The techniques described in the above-mentioned Finnish Patent Applications FI20055252, FI20055253 and FI20055254 can be used to process spectral information that is transported in the search for individual lipids.

The techniques described in the aforementioned Finnish Patent Application FI20055198 can be used to combine lipid compound information with boy chart information as well as information at other levels of biology.

Figure 12 shows a data representation showing how inter-tissue lipid profiling at a single molecular time interval can reveal interdependencies between biological processes in different compartments of organisms. Data 1200 shows a significant association of cardiac LysoPC (lysophosphate dimethylcholine) with 20 liver TAGs (triacylglycerol) and a negative association with brown adipose tissue (BAT) GPEtnm (glycerylphospho-ethanolamine). For example, the addition of specific triacylglycerols in the liver is associated with the addition of ether-linked lysophosphate dylcholine in the myocardium, which is linked to mitochondrial dysfunction in the heart. See.

25 eg co-regulation between elements TAG 54: 3 and LysoPC 16: 1e. This one

the code control, designated 1202, is a surprising discovery made by the method and data processing system of the invention. |

Finally, Figure 13 summarizes the steps of the method of the invention.

30 One skilled in the art will readily appreciate that as technology advances, the inventive concept can be implemented in many different ways. The invention and its embodiments are thus not limited to the examples described above, but may vary within the scope of the claims.

! |

Claims (8)

A method for processing information on compounds of moiety molecules which share common moieties, comprising: maintaining path diagram information of compounds at the level of the individual compound and / or at the generic class level (13-1); developing a diversity of compounds based on a plurality of seed structures, each of which describes a higher-than-average lipid compound in nature (13-2); using a formal description language to express seed structures 10 (13-3); using the structural elements to generate the expected spectra for each compound using known experimental conditions for mass spectrometry (13-4); performing one or more spectroscopy experiments to obtain compound information (13-5); and combining the resulting compound information with existing information on molecular classes (13-6). The method of claim 1, wherein the existing information on the compounds includes path diagram information at the level of the individual compound and / or at the generic class level. The method of claim 1 or 2, wherein the existing compound information includes co-regulatory information with other compound information from different biological samples. The method of any preceding claim, further comprising linking information from the individual compound to information at other levels. The method of claim 4, wherein the information at other levels comprises information about proteins or genes associated with the metabolism or biological variation of the individual compound. The method of any one of the preceding claims, further comprising using information at the level of the individual compounds and their variation within a particular portion of the organism in different biological samples to find interdependencies between the different parts of the organism. The method of any one of the preceding claims, wherein the compounds of the molecular classes comprise lipids. A pot processing system for processing information on compounds of molecular classes sharing common components, the data processing system comprising: a database for maintaining off-chart information on compounds at the level of the individual compound and / or at the generic class level; and processing genetics: to develop diversity of compounds based on a set of seed structures, each representing a lipid compound that is more likely to occur in nature than average; 10. use a formal description language to express seed structures; use structural elements to generate expected spectra for each compound using known experimental conditions for mass spectrometry; - perform one or more spectroscopy experiments to obtain compound information; and - linking the obtained compound information to existing information on molecular classes.